Google's Keyword Planner returned 6,256 home project searches for our example market, Raleigh-Durham. That sounds like a lot of keywords to buy, until you look at the sizes.
Only 108 of them show 100 or more searches a month. Another 4,233 show somewhere between 10 and 90. The last 1,915 are too small for Keyword Planner to put a number on at all.
Scroll to the bottom of that list and you find the most honest searches in the whole industry. "I want to convert my garage into living space." "Adding a partial second story to a ranch house." "Attached garage addition with living space above." "Converting a garage into a bedroom and bathroom." Each one shows about 10 searches a month, and Keyword Planner has no bid data for them, which means almost nobody is paying to show up for them.
Those searches are a homeowner describing their own house. No contractor would ever think to type them into a keyword list, and that's exactly the problem this chapter solves. It covers how we split a contractor's account into two campaigns that do two different jobs: one that owns the searches you can name, and one that goes looking for the ones you can't.
Exact Match Belongs to the Kitchen and Bath Searches You Can Already Name.
The first campaign is the one we've been building since Chapter 1. We call it Isolation. It holds Chapter 5's project ad groups, every keyword on exact match, every bid typed in by a person. "Kitchen remodel contractor." "Bathroom remodel near me." "Home addition contractor." The searches every remodeler in your market knows about.
Exact match is the tightest leash Google offers. Think of it as a short leash. Per Google, an exact keyword only triggers your ad on a search with the same meaning, so you steer hard and give up reach in return.
Less reach sounds like a weakness. For a new account, it's the point. Isolation spends on the searches most likely to turn into a booked estimate, so the account starts teaching Google what a real lead looks like from the first week. Chapter 4 explained that Quality Score readings build on exact searches. Chapter 1 set the booked estimate as the conversion Google bids toward. Isolation is where both of those start paying off.
The 108 is smaller than it looks, too. Keyword Planner lists close variants as separate rows with the same number: "bathroom remodel," "bathroom reconstruction" and "bathroom remodel bathtub" all show 1,000 in the example market, and "kitchen remodel contractors," "kitchen renovation contractors" and "contractor for kitchen remodel" all show 590. Collapse those and a day-one Isolation list for a kitchen, bath and addition remodeler in that market comes down to a few dozen exact keywords across five or six ad groups, led by:
- Baths: "bathroom remodel" (about 1,000 a month), "bathroom remodel contractors" (880), "bathroom remodel near me" (480)
- Kitchens: "kitchen remodel contractors" (590), "kitchen remodel" and "kitchen remodel near me" (480 each)
- Whole-home and additions: "home renovation contractors" (390), "home addition contractor" (260)
- Outdoor living, if you build it: "deck contractors near me" (720)
A list that short is a feature. In month one, the budget only buys searches you'd be proud to answer.
What Isolation can't do is find the homeowner who typed "converting an attached garage into living space." Your list doesn't have it. Nobody's list has it.
A Homeowner Planning a Garage Conversion Doesn't Type What Your Keyword List Says.
People don't search like contractors talk. A contractor says "garage conversion." A homeowner types the whole situation: the one-car garage, the studio apartment, the granny flat.
Look at what those ten searches tell you. Five of them mention a garage. Two are about building up instead of out. Most are about adding living space without moving: a bedroom and bath over the garage, a partial second story, a finished basement with a bathroom. Several name the exact layout they want, and one asks outright for contractors. Not one of them would show up in a list built from "general contractor" and "kitchen remodel."
That's the long tail: thousands of small, specific searches, each too small to plan around. In our experience, together they add up to a large part of the people actually shopping for a remodel, and they're where a contractor's account grows once the basics work, because the competition hasn't arrived yet. You can't write that list in advance. You have to let Google find it, carefully, inside a fence you control.
Broad Match Finds Them Once Exact Match Has Booked Enough Estimates to Teach It.
The second campaign is Exploration (also our name), and it does what Isolation can't. We build it in four moves:
- Duplicate the Isolation campaign.
- Delete every keyword except the handful, three to five, that already land on your estimate calendar week after week.
- Switch them to broad match.
- Hand the bids to Google, aimed at booked estimates, with a ceiling: a target cost per estimate copied from Isolation's real results. That ceiling is the guardrail.
Broad match reads a keyword as an idea, not a phrase. Google says broad match reads three extra clues to guess what a homeowner means: what that person searched for a little earlier, what your landing page and ad say, and which other keywords share the ad group. On the pairing, Google doesn't hedge: it calls automated bidding alongside broad match critical.
That's why Exploration waits. Broad match with a bidding system that has nothing to learn from is a contractor handing an open purchase order to a stranger. Our trigger for Exploration is volume: roughly 30 quality conversions in a month. For most contractor accounts that lands around months 2-3. That's a typical calendar, not a promise; a remodeler with a bigger budget and a busier phone hits 30 sooner. Give it a month or two to settle in.
Setting that target is where most contractors get greedy. Google's own rule of thumb, from its target-CPA help page: base the first target on last month's real cost per conversion, then don't grade it until 30-plus conversions have come through. Google also warns that "Setting a target that's too low... may cause you to forgo clicks that could result in conversions." Say your hand-bid campaign has been landing estimates for roughly $350 apiece. Launch Exploration at about that number. The $200 you'd like to pay is a goal, not a starting bid. Tighten it later, in small steps, once Exploration has a month of its own numbers.
The first 30 days of Exploration are the most hands-on part of the whole rollout. Each morning, somebody pulls up the search terms report (the real wording homeowners typed before your ad appeared) and works it like a punch list:
- Blocks the junk. A permit question from a homeowner planning to do the work themselves. "Contractor license class." A town you don't serve. Each goes on the negative list that day, alongside the shared list Chapter 5 already built.
- Flags the winners. Any search that books an estimate gets noted for promotion.
After the first month, the checks can drop to a few times a week. They never stop. The report has a blind spot. In Google's words, "Some search terms that don't have enough query activity are omitted from the search terms report in order to keep with our standards on data privacy." You'll never see every search. You'll see enough to steer.
What if the account never reaches 30 conversions a month? Plenty of smaller contractors won't, at least not at first, and that's fine. Isolation keeps working on its own. Read its search terms report too: exact match still shows searches with the same meaning as your keywords, and the good ones can be added as exact keywords by hand. That's a slower way to grow the list, but it never hands the budget to a bidding system that has too little to learn from.
Exact Match Holds the Ground You Already Won. Broad Match Pushes the Fence Line Out.
The two campaigns aren't a phase one and a phase two. They work together for as long as the account runs, and the balance between them shifts over time.
Promotion. When an Exploration search books estimates, it moves into the house. It gets its own exact match keyword in the matching project ad group back in Isolation, and its own hand-set bid. Now you control exactly what you pay for it, and Exploration is free to go find the next one.
Pausing and reclaiming. There may come a stretch when Exploration books most of the estimates at a price you like; Isolation can sit idle then. If one of your money searches starts sliding down the page, switch Isolation on for it again and buy the spot back at a price you choose. Chapter 7 covers the columns that tell you when a search is slipping, and how often to look at them.
The fence. The negative list is the property line. It keeps broad match from wandering into job seekers, do-it-yourselfers, single trades you don't do and commercial work. It needs mending as often as any real fence.
What you never do is skip Isolation and launch on broad match because it "reaches more people." It does. It also reaches people looking for contractor jobs, people pricing a project they'll do themselves and people shopping for a trade you don't offer. Exact match first is how a new account learns what a good lead is before it goes looking for more of them.
If you'd rather have both campaigns run for you, our Google Ads management builds every contractor account in exactly this order.
Next: A Bathroom Click and an Addition Click Shouldn't Carry the Same Bid.
Chapter 7 sets the bids for Isolation one project at a time, reads the columns that show whether you're winning the searches that matter, and turns Isolation's real cost per estimate into the target Exploration uses.
This chapter is part of Google Ads for Contractors, our sixteen-chapter guide for residential general contractors and construction companies anywhere in the US.
